Peptic Ulcer Disease
002.5-7125/91 $0.00 + .20
Surgical Treatment of Peptic
Ulcer Disease
Ajit K. Sachdeva, MD, FRCS(C), FACS,*
Howard A. Zaren, MD, FACS,t
and Bernard Sige/, MD, FACS:!:
Surgery for peptic ulcer disease has gone through a period of significant
change over the past 15 years. The declining incidence of the disease, the
introduction of H 2 -receptor antagonists (as well as other effective drugs to
control acid secretion), the development of proximal gastric vagotomy, and
re-evaluation of older procedures have all contributed to the emergence of
a new era in the surgical management of peptic ulcer disease.
The incidence of peptic ulcer disease was on the decline even prior to
the introduction of H 2 -antagonists in 1977, and the decline in the number
of patients hospitalized for uncomplicated peptic ulcer disease has continued
since that time. However, hospitalization rates for hemorrhage and perforation have remained constant. 33 In some series, hospital admissions for
hemorrhage have even shown an increase. 1 Operations for peptic ulcer
disease have followed these general trends. Emergency procedures for
hemorrhage have increased and account for from 20% to 50% of all surgical
procedures performed for peptic ulcer disease. 22,38, 34
.
The use of proximal gastric vagotomy without a drainage procedure for
the treatment of peptic ulcer disease was first reported between 1969 and
1970. Since that time, this "physiologic" procedure increased in popularity
and then because of the significant ulcer recurrence rate enthusiasm
diminished. However, the p;'ocedure is still a viable option in several
clinical situations because of its low complication rate.
Several age and gender differences in patients with peptic ulcer disease
are worthy of note. In recent years, patients requiring surgery for peptic
*Associate
Professor of Surgery, and Director of Surgical Education, Medical College of
Pennsylvania; and Chief, Surgical Services, Veterans AfFairs :\ledical Center, Philadelphia,
Pennsvlvania
tProfessor ~f Surgery, and Interim Chairman, Medical College of Pennsylvania, Philadelphia,
Pennsvlvania
:j:Professor '01' Surgery, and Director of Surgical Research, J\[edical College of Pennsylvania,
Philadelphia, Pennsylvania
Medical Clinics of North America-Vol. 7.5, No. 4, July 1991
999
lOOO
AJIT
K.
SACHDEVA ET AL.
ulcer disease have been older and have associated medical problems.
Although the majority of candidates for surgery are men, the proportion of
women requiring such operations has increased. 22 The prevalence of nonsteroidal anti-inflammatory drug usage in the adult population might account
for the increase seen in hospitalizations for bleeding gastric ulcers. 33
The operative mortality following elective surgery is generally from
1% to 2% and depends on the type of procedure as well as the operative
risk of the patient. Emergency operations are associated with an almost 10fold increase in operative mortality.22 This underscores the need for earlier
identification of patients who might benefit from elective surgery for peptic
ulcer disease.
OPERATIONS FOR PEPTIC ULCER DISEASE
The surgical procedure for peptic ulcer disease must be taiiored to the
specific needs of the individual patient. Efficacy of the procedure must be
balanced against the risk to the patient. For optimum results, the surgeon
should be able to select the appropriate procedure from a variety of
available options.
Truncal Vagotomy and Drainage
Vagotomy remains pivotal in the surgical treatment of duodenal ulcer
disease. Vagotomy decreases acid production by diminishing the cholinergic
stimulation of the parietal cells and by decreasing the response of parietal
cells to gastrin. Reports indicate that basal acid production is reduced by
70% and stimulated acid production is reduced by 50% following truncal
vagotomy.40 Due to total denervation of the stomach (along with other intraabdominal viscera), truncal vagotomy alone would result in poor gastric
emptying and gastric stasis. Hence, this procedure must be combined with
a drainage procedure, either pyloroplasty (Fig. 1) or gastrojejunostomy.
Truncal vagotomy and drainage are relatively easy to perform and are
associated with low operative mortality and with an ulcer recurrence rate
of from 10% to 15% (Table 1). Mild symptoms of diarrhea and dumping are
fairly common following this procedure, but they can be managed conser-
Figure 1. Truncal vagotomy and Heineke-Mikulicz pyloroplasty.
1001
SURGICAL TREATMENT OF PEPTIC ULCER DISEASE
Table 1. Results of Operations for Duodenal Ulcer Disease
OPERATIVE
ULCER
PROCEDURE
MORTALITY
RECURRENCE
Truncal vagotomy and drainage
Truncal vagotomy and antrectomy
Subtotal gastrectomy
Proximal gastric vagotomy
1%
1-2%
1-2%
<O.S%
1O-1S%
<1%
3-S%
1O-1S%
Data from references 12, 2S, 40, and SO.
vatively in most cases. The incidence of severe diarrhea and dumping is
relatively low (Table 2).
Truncal Vagotomy and Antrectomy
The addition of antrectomy to truncal vagotomy (Fig. 2) results in the
most effective operation for duodenal ulcer disease. Using this surgical
treatment, the basal acid production is reduced by 85% and stimulated acid
production is reduced by 80%.40 Reconstruction may be performed by a
gastroduodenostomy (Billroth I) or a gastrojejunostomy (Billroth II). Although truncal vagotomy-antrectomy is associated with a slightly higher
mortality rate, in elective situations with good-risk patients, mortality rates
have been quite low. The ulcer recurrence rate for the truncal vagotomyantrectomy procedure is very low (see Table 1), and the incidence of
diarrhea and dumping is similar to truncal vagotomy and drainage procedures (Table 2).
Subtotal Gastrectomy
Subtotal gastrectomy without vagotomy involves excision of up to 75%
of the distal stomach with a Billroth I or Billroth II anastomosis. This
procedure results in the removal of the gastrin-producing antrum and part
of the parietal cell mass. The operative mortality rates are generally in the
same range as for truncal vagotomy-antrectomy. However, the ulcer recurrence rate after subtotal gastrectomy is higher than for truncal vagotomyantrectomy (see Table 1). Although the incidence of diarrhea after subtotal
gastrectomy has been reported to be a little lower than for vagotomyantrectomy in some series, the incidence of postoperative dumping is
similar for both procedures. 12
Table 2. Undesirable Consequences Following Operations
for Duodenal Ulcer Disease
DIARRHEA
PROCEDURE
Truncal vagotomy and
drainage
Truncal vagotomy and
antrectomy
Proximal gastric vagotomy
DUMPING
Mild
Severe
Mild
Severe
20-2S%
1%
10-20%
1% (or higher)
20-2S%
1%
10-20%
1% (or higher)
<S%
1%
Data from references 23, 40, SO, and SS.
2%
<1%
o
~
~
~
Figure 2. Truncal vagotomy and antrectomy.
SURGICAL TREATMENT OF PEPTIC ULCEH DISEASE
1003
Proximal Gastric Vagotomy
Proximal gastric vagotomy involves division of the vagus nerve branches
to the parietal cells, leaving the nerve supply to the antrum and pylorus
intact. Hence, the nerves of Latarjet are preserved and the proximal
stomach denervated, including the distal 6 cm of the esophagus. The nerves
are divided up to a point approximately 5 to 6 cm proximal to the pylorus
(Fig. 3).24 Acid production may be reduced 80% to 90% in the early
postoperative period, but this usually diminishes to 70% to 80% and remains
at that level for years. Stimulated acid production may be reduced as much
as 80% soon after the operation. However, this reduction also drops to
around 50% after 1 year. 44 Although proximal gastric vagotomy is technically
more demanding and is difficult to perform in obese individuals, it has
been associated with a very low operative mortality (see Table 1). The
operation does not require a drainage procedure, and other undesirable
consequences are fewer (Table 2). However, the high ulcer recurrence rate
reported following this procedure has made the procedure less popular.
Recurrence rates of up to 30% have been reported in the literature, but
the usual recurrence rates are around 10% to 15% when surgical technique
is meticulous and includes denervation of the distal esophageal region
(Table 1). Higher recurrence rates have also been reported in some series
when the procedure was performed for pyloric channel or prepyloric ulcers.
Selective Gastric Vagotomy
In an attempt to selectively denervate the entire stomach (including
the antrum and pylorus) but preserve the hepatic and celiac branches of
the vagus nerves, the procedure of selective gastric vagotomy was performed
by several surgeons in the United States and in Europe. However, it does
require a drainage procedure because the antrum and pylorus are denervated. A recurrence rate of 2% after 5 years has been reported, attesting
to the efficacy of selective gastric vagotomy.29 A lower incidence of diarrhea
compared with truncal vagotomy has been reported in a few series, but the
incidence of dumping after selective gastric vagotomy is similar to that
following truncal vagotomy and drainage. 12 Overall, the procedure does not
have significant advantages over truncal vagotomy and drainage, and it is
Figure 3. Proximal gastric vagotomy.
1004
AJIT K.
SACHDEVA ET AL.
technically more demanding. Hence, selective gastric vagotomy currently
is not very popular.
Other Complications Following Operations for Peptic Ulcer Disease
In addition to diarrhea and dumping, another troublesome complication
following surgery for peptic ulcer disease is alkaline reflux gastritis. Alkaline
reflux gastritis may occur after gastric resections (with Billroth I or 11
reconstruction) or following pyloroplasty. However, it is seen very rarely
following proximal gastric vagotomy. If the diagnosis of alkaline reflux
gastritis is confirmed by endoscopy and biopsy in a patient who presents
with epigastric pain and bilious vomiting, reoperation may be indicated in
severe cases. In cases in which the original procedure was resection and
Billroth I or 11 anastomosis, conversion to a Roux-en-Y anastomosis may be
undertaken (Fig. 4). If the original procedure was vagotomy and drainage,
resection with Roux-en-Y reconstruction should be performed. Vagotomy
must be added, if not previously performed, to prevent subsequent ulceration with a Roux-en-Y procedure. Results are generally favorable following
Roux-en-Y conversion, although failure rates as high as 30% to 50% have
been reported. 41 Also, Roux-en-Y conversion may result in delayed gastric
emptying;
Several authors have reported increased incidence of carcinoma in the
gastric remnant 15 to 25 years after partial gastrectomy for benign peptic
ulcer disease. In one series, risk appeared to be 3.2 times higher after 25
years, as compared with the general population. 52 Hence, careful endoscopic
surveillance of such patients is indicated 12 to 15 years following gastrectomy.
Other complications of gastric surgery include afferent loop syndrome,
bezoar formation, and malnutrition.
Figure 4. Roux-en-Y reconstruction after partial or
subtotal gastrectomy.
SURGICAL TREATMENT OF PEPTIC ULCER DISEASE
1005
INDICATIONS FOR SURGERY AND CHOICE OF OPERATION
FOR PEPTIC ULCER DISEASE
Hemorrhage
The overall mortality for hemorrhage from peptic ulcer disease remains
around 10% to 20%. Although 75% of patients admitted with massive
hemorrhage from peptic ulcer disease will stop bleeding spontaneously
within 48 hours, some patients with life-threatening hemorrhage will
require emergency surgery. Mortality rates are higher for patients who are
older than 60 years, for patients who require more than five units of blood
transfusion, and for patients who have multiple system failure. 35 Hunt20
reported that shock on admission to the hospital was associated with a
significantly greater incidence of re-bleeding in 70% of cases. Also, patients
with endoscopic stigmata of recent hemorrhage and those with hemoglobin
levels of less than 8 g/dL at admission appear to be more likely to have
further bleeding. 7 A visible vessel on endoscopy has been reported to be
associated with up to a 60% chance of further bleeding. 27 Chronic ulcers
that are bleeding from an artery at the time of endoscopy continue to bleed
or re-bleed in 80% to 100% of patients and represent a very high-risk
group.36 Patients with gastric ulcers have a greater risk of re-bleeding than
patients with duodenal ulcers.5 An aggressive approach in these high-risk
groups is indicated in order to decrease mortality.
The use of endoscopic methods to control hemorrhage may be attempted as the first therapeutic step. These methods include monopolar or
bipolar electrocoagulation, laser photocoagulation, and direct application of
heat with a "heater" probe. The results of these treatment modalities have
been mixed and are clearly operator-dependent. 13, 28, 36, 48 If these methods
are unsuccessful or if brisk bleeding precludes the use of these methods,
emergency operation must be performed immediately because delay in
surgical intervention results in higher patient mortality,43 It should also be
pointed out that on occasion, if bleeding is brisk, it may not be possible to
adequately resuscitate the patient prior to the operation. In such cases, the
patient must be taken to the operating room expeditiously, even if vital
signs continue to remain unstable; otherwise, the condition of the patient
will continue to worsen, resulting in increasing transfusion requirements
and a formidable risk of perioperative mortality.
Once the decision to operate has been made, it is very important to
gain control of the bleeding vessel as quickly as possible. For a bleeding
duodenal ulcer, an anterior duodenotomy is performed with suture ligature
of the vessel at the base of the ulcer. Following this step, the choice of
operation depends on several factors: location of the ulcer; age, stability,
and general medical condition of the patient; and preference of the
individual surgeon. For older patients and for those who are hemodynamically unstable, we prefer a truncal vagotomy-pyloroplasty. This procedure
can be performed expeditiously and has a lower mortality rate than
procedures involVing gastric resection. In younger, lower-risk patients, we
recommend truncal vagotomy-antrectomy, with a Billroth I or 11 reconstruction. However, a very scarred duodenum might make this procedure
hazardous due to the difficulty in dealing with the duodenal stump. A
1006
AJIT
K.
SAClIDEYA ET AL.
Billroth II procedure might thcn requirc special steps, such as a tube
duodenostoll1Y, or closure of the duodenal stump with a RO\lx-en- Y loop of
jejunum. A scarred anterior duodenal wall might even make a pyloroplasty
technically very difficult to perform. Although proximal gastric vagotomy
has been successfully perf()rmed in patients requiring emergency surgery
f()r massive hemorrhage,IO, 17 we do not recommend this approach because
it prolongs operating time in these critically ill patients. If proximal gastric
vagotomy is selected, the patient should be younger, lower-risk, and
hemodynamically stable. Thc incidence of early re-bleeding following
truncal vagotomy-pyloroplasty has been reported to be 4.3% versus a ratc
of 0 ..5% for early re-bleeding following truncal vagotomy-antrectomy.16
For bleeding gastric ulcers located distally, a distal gastrectomy with a
Billroth I or 11 reconstruction should be performed in relatively good-risk,
stable patients. Since the resected specimen includes the ulcer, this also
allows adequate pathologic evaluation of the ulcer. In higher-risk patients,
excision of the ulcer with vagotomy-pyloroplasty may be performed. j(j For
gastric ulcers located high along the lesser curve, following control of
hemorrhage, biopsy of ulcer should be performed along with vagotomypyloroplasty. In very unstable patients, simple suture ligature of the
bleeding vessel and biopsy of the ulcer should be considered. 42
Stress Ulceration
Stress ulceration is associatcd with a number of risk factors, including
sepsis, trauma, shock, and multiple-system failure. Prophylaxis of stress
ulceration in these situations (with antacids, Hz-receptor antagonists, proton
pump inhibitors, or sucralhtte) and correction of the associated predisposing
condition remain the primary methods of patient management. Such
regimens have been successfiJl in preventing blecding in 88% to 97% of
cases. 56 However, if hemorrhage occurs, and if it is of sufficient magnitude
to requirc surgery, mortality rates of 40% to 60% result. Factors contril)uting to high mortality include multiple predisposing causes, massive blood
transfusions (17 units or more), rcspiratory failure, and recurrent hemorrhage. 19 When hemorrhage persists, operation should be considercd after
transfusion of five to six units of blood.
Operation for stress ulceration must be adequate to diminish the
incidence of continued or recurrent bleeding that results in high mortality
rates. Also, mortality rates correlate more accurately with the general
condition of the patient than with the operative procedure performed.
Hence, a more aggressive surgical approach is indicated. We recommend
a near-total or total gastrectomy for diffuse mucosal ulceration and bleeding.
For the few patients with limited distal gastric ulceration, a distal gastrectomy may be adequate. Oespite initial control of hemorrhage following
surgery, a significant number of patients are likely to re-bleed. In a
retrospective review covering a period of 25 years, Hubert and associates 19
reported recurrent bleeding in 38% of cases following surgery.
In poor-risk patients, emergency arteriography and selective embolization might be considered instead of surgery as the initial interventional
approach. C sing these techniques, control of hem or rh age has been reported
in 79% of cases and recurrent bleeding has been documented in 18% of
SUHCICAL TREAT\IE'JT OF PEPTIC ULCEH DISEASE
1007
cases. 9 However, the complication rate for this technique remains quite
high, even in experienced hands.
Perforation
If patients with perforated duodenal ulcer are in shock, they require
aggressive resuscitation followed by prompt celiotomy. Preoperatively,
nasogastric aspiration is instituted and broad-spectrum antibiotics started.
There is still some controversy as to whether these patients should
undergo only closure of the perforation with an omental patch or whether
a definitive ulcer operation should be combined with closure of perforation.
It has been claimed that a prior history of duodenal ulcer disease or findings
of chronic ulcer at the time of the operation might identify suitable patients
who will bencfit from immediate definitive surgery. However, this approach
has been challenged because of the problems associated with obtaining an
accurate history and because of the high ulcer recurrence rates reported in
some series, irrespective of the duration of ulcer symptoms. In a series
reported by Boey and associates, ulcer recurrence rates following simple
closure of perforation were 36.6% after 3 years, compared with a 10.6%
recurrence rate after definitive surgery (proximal gastric vagotomy). Recurrence rates between 14% and 80% f()lIowing closure of acute ulcers have
been reported.:] Authors have reported comparable operative mortality in
relatively good-risk patients whether simple closure is performed or a
definitive procedure is added. 6 This makes a strong case for the latter
approach in patients with perforated duodenal ulcer. 14 Despite the recommendations of some authors who suggest just closure of the perforation, 4. 30
we recommend a definitive procedure unless the patient is poor-risk and
has significant peritoneal contamination or unless the peIforation is associated with an acute ulcer caused by drug ingestion or acute stress.
The choice of the definitive operation also remains somewhat controversial. In patients with duodenal ulcer perforation, good results have been
reported with proximal gastric vagotomy. 3. 21 Alternatively, truncal vagotomy-pyloroplasty may be peIformed. We have used truncal vagotomypyloroplasty quite frequently because its technical ease reduces operating
time considerably.
Perioperative mortality in patients with perforated duodenal ulcer has
been found to be related to preoperative shock, severe concurrent medical
illness, and presentation for treatment 24 hours after peIforation. 2 If all
three risk factors were present, Boey and associates 2 reported mortality of
100%. Koness and associates 31 reported that patients with perforated ulcer
who were older than 5,5 years and those with intraoperative hypotension
had a higher mortality.
For perforated gastric ulcers, a distal gastrectomy with Billroth I or 11
anastomosis should be undertaken if the ulcer is in the distal stomach and
the patient is a good risk. .39 Unlike duodenal ulcers, gastric ulcers present
the risk of malignancy. Hence, this procedure has the additional advantage
of enabling excision of the ulcer for pathologic diagnosis. In high-risk
patients, biopsy of the ulcer should be performed and the ulcer closed with
an omental patch. 34 ..51, 53
1008
AJIT K.
SACHDEVA ET AL.
Obstruction
Patients with peptic ulcer disease may present with acute gastric outlet
obstruction because of inflammation and edema around a pyloric channel
ulcer. Other patients present with a chronic history of ulcer disease that
has resulted in scarring of the pyloroduodenal region and thus permanent
narrowing. Many patients with permanent narrowing have a long history
of ulcer disease, with periodic exacerbations.
When thc diagnosis of gastric outlet obstruction is made, the patient
should be placed on a regimen of nasogastric decompression, intravenous
fluid and electrolyte replacement, and Hz-receptor antagonists. In patients
with acute obstructions, improvement is generally seen within 72 hours. If
the patient does not improve, operation should be performed. A slightly
longer period of preoperative management may be required if the patient
has been significantly depleted of fluid and electrolytes (resulting in severe
metabolic problems), or if the patient needs parenteral nutritional support.
In younger, better-risk patients, we recommend truncal vagotomyantrectomy, with a Billroth 11 reconstruction. This procedure is very
effective and has a low ulcer recurrence rate. In older patients, truncal
vagotomy and a drainage procedure should be considered. Delayed gastric
emptying following operative procedures has been reported by some
authors, but has not been found to be a significant problem by others. 11
Adequate preoperative nasogastric decompression helps to diminish gastric
atony and facilitates postoperative gastric emptying.
Several recent reports note good results with proximal gastric vagotomy
and drainage. b . .37 Also, proximal gastric vagotomy with dilatation has been
performed in patients with gastric outlet obstruction. 26 However, the longterm results of dilatation are still questionable. Kozarepz reported immediate symptomatic relief in 67% of patients following endoscopic dilatation.
However, objective improvement (radiologically or endoscopically) at 3
months was seen in only 38% of patients. Restricturing after dilatation
(surgical or endoscopic) continues to remain a concern.
Intractability
The availability of effective medication has led to high rates of healing
of duodenal ulcers; for example, it is stated that over 80% of patients heal
after 4 weeks of starting medical therapy and over 90% of patients heal
after 8 weeks. 45 However, once treatment is stopped, 50% to 90% of
patients will have a recurrence. 46 Thus, long-term maintenance therapy
with H 2 -receptor antagonists is recommended. Medication does not, however, cure the underlying ulcer diathesis, and about 10% of patients will
ultimately develop complications (hemorrhage, perforation, or stenosis).46
In some patients, ulcer recurrences are frequent, or ulcers do not respond
to the usual medical management. These patients should be considered
intractable and therefore candidates for elective surgery.40 Persistence with
long-term medical treatment in these cases may result in a high complication
rate. Also, a few patients will continue to be noncompliant with medical
therapy, and they too should be considered candidates for elective surgical
treatment. The cost of long-term maintenance therapy in comparison to the
cost of surgery must be considered during treatment of duodenal ulcer
SURCICAL TI\EAT\IE]\;T OF PEPTIC CLCEI\ DISL\SE
1009
disease. 'Neighing all these f~lctors, if electivc surgery for duodenal ulcer
disease is indicated, we recommcnd proximal gastric vagotomy because of
its very low complication rate coupled with a very low incidence of
undesirable postoperative side effects (such as dumping or diarrheal. If
proximal gastric vagotomy is not advisable for technical reasons, another
procedure may be selected.
Prior to initiation of medical therapy, a gastric ulcer must be evaluated
endoscopically and biopsy perfcmned to assess for possible malignancy.
Once medical treatment is begun, close follow-up is required. If the gastric
ulcer does not heal within 6 to 8 weeks of placing the patient on medical
therapy, we recommend elective surgery. For distal gastric ulcers, a distal
gastrectomy should be performed. .
.
Recurrent Ulceration
Recurrent ulceration may occur after a surgical procedure for peptic
ulcer disease. The incidence varies according to the type of procedure
performed and may be a result of technical error during the original
procedure. Examples of technical errors resulting in recurrent ulceration
include incomplete vagotomy, inadequate gastric resection, and retained
gastric antrum in the duodenal stump following a Billroth 11 reconstruction.
If recurrence occurs, other causes such as Zollinger-Ellison syndrome and
hyperparathyroidism must also be considered. The recurrence rate after
operations for benign gastric ulcers has been reported to be between 2%
and 4%Y
Once Zollinger-Ellison syndrome and hyperparathyroidism have been
excluded, the patient may be placed initially on medical treatment. Hoffmann reported an 80% rate of healing of recurrent ulcers in patients on
H 2 -receptor antagonist therapy after 4 weeks and a 95% rate of healing
after 12 weeks of therapy. However, 30% to 75% of patients will develop
a new ulcer within 9 months of stopping this treatment. Even on maintenance therapy, up to 50% of patients will develop ulcer recurrence.
Generally, 50% of the patients who develop recurrent ulceration will
require a second operation. 1R
A variety of surgical options must be considered, depending upon the
type of original procedure that was performed, the operative findings at
the time of that first procedure, the pathology reports fi'om the first
procedure, and other diagnostic studies that might provide information on
a possible technical error that may have contributed to the recurrence. If
the initial procedure was vagotomy and drainage or proximal gastric
vagotomy, re-vagotomy and antrectomy should be considered. If the initial
procedure was vagotomy-antrectomy, re-vagotomy or further gastric resection may have to be performed.1.5 If there is stenosis of the gastroenteric
anastomosis, this should be revised at the time of reoperation. If there is
no reason to go into the abdomen other than to perform re-vagotomy, it is
technically much easier and safer to perform the vagotomy through the
transthoracic route, away from adhesions from previous surgery.49 For
recurrent gastric ulceration, further gastric resection may need to be
performed. The specific operative procedure for recurrent ulceration depends on the requirements of the individual patient.
1010
AJIT
K.
SACHDEYA ET AL.
SUMMARY
Elective surgery for peptic ulcer disease has diminished significantly
over the past 15 years. However, emergency surgery has not shown a
decline. Some series have even reported an increase in hospitalizations and
operations for hemorrhage. The appropriate surgical procedure for peptic
ulcer disease must be tailored to the specific needs of the individual patient.
During emergency operations for hemorrhage from duodenal ulcer,
we recommend suture ligature of the bleeding vessel and vagotomypyloroplasty for high-risk patients, or vagotomy-antrectomy for the lowerrisk patient. Bleeding gastric ulcers should be resected, if possible. For
massive hemorrhage from stress ulceration requiring surgery, near-total or
total gastrectomy should be performed.
Perforated duodenal ulcers are best managed by closure and a definitiye
ulcer operation, such as vagotomy-pyloroplasty. Perforated gastric ulcers
are best excised but may be simply closed if conditions do not favor
resection. In these situations, biopsy should be performed.
\Ve recommend truncal vagotomy-antrectomy for patients presenting
with obstruction. Vagotomy (truncal or proximal gastric) with drainage is
an acceptable alternative in this situation.
For patients with intractable ulcer disease or for those who are
noncompliant, proximal gastric vagotomy is the preferred operation. Howeyer, other operations may need to be considered, depending on the
specific situation.
Recurrent ulceration needs appropriate work-up to determine the
possible cause. Although patients with ulcer recurrence initially may be
placed on medical treatment, about 50% will require reoperation.
The most effective procedure for peptic ulcer disease is truncal
vagotomy-antrectomy, which has a recurrence rate of less than 1%. The
procedure with the least morbidity and the fewest undesirable side effects
is proximal gastric vagotomy. Ulcer recurrence after proximal gastric
vagotomy or truncal vagotomy-pyloroplasty is in the range of 10% to 15%.
ACKNOWLEDCl\IEKTS
The authors wish to acknowledge the contributions of Juliette K. Smith, who prepared
the illustrations for Figures 1 through 4, and Maureen Benzing, who assisted in preparation
of the manuscript and provided technical editing.
REFERENCES
1. Bardhan KD, Cust C, HinchlifI"e RFC, et a1: Changing pattern of admissions and operations
for duodenal ulcer. Br J Surg 76:230, 1989
2. Boey J, Choi SKY, Poon A, et al: Risk stratification in perforated duodenal ulcers: A
prospective validation of predictive factors. Ann Surg 205:22, 1987
3. Boey J, Branicki FJ, Alagaratnam TT, et al: Proximal gastric vagotomy: The preferred
operation ft)r perf()rations in acute duodenal ulcer. Ann Surg 208:169, 1988
4. Bormnan PC, Theodorou NA, Jeflery PC, et a1: Simple closure of perforated duodenal
ulcer: A prospective evaluation of a conservative management policy. Br J Surg 77:73,
1990
SCI\CIC\L THKH"!E".;T OF PEPTIC l'LCEH DISEASE
1011
5. Brearlev S, Hawker PC, ~lorris DL, et al: Selection of patients f()r surgerv f()llowing
pepti~· ulcer haemorrhage. Br J Surg 74:H9:3, 19H7
6. Christiansen J, Andersen OB, Bonnesen 1', et al: Perf(m,ted duodenal ulcer managed bv
simple closure versus closure and proximal gastric vagotomv, Br J Surg 74:286, 19H7
" Clason AE, Macleod DAD, Elton RA: Clinical hlctorS in the prediction of hlrther
haemorrhage or mortality in acute upper gastrointestinal haemorrhage. Br J Surg
73:985, 19H6
8. Donahue PE, Yoshida J, Richter HM, et al: Proximal gastric vagotomy with drainage ft)r
obstructing lluodenal ulcer. Surgery 104:7.57, 19H8
9, Eckstein 1\1R, Kelemouridis V, Athanasoulis CA, et al: Gastric bleeding: Therapy with
intraarterial vasopressin and transcatheter emholization, Hadiology 1.52:643, 1984
10, Falk CL, Hollinshead JW, Cillett DJ: Highly selective vagotomy in the treatment of
complicated duodenal ulcer. Med J Austr 152:574, 1990
11. Fisher HD, Ebert PA, Zuidema CD: Obstructing peptic ulcers. Arch Snrg 94:724, 1967
12, Fromm D: Ulceration of the stomach and duodenum, In Fromm D (ed): Castrointestinal
Surgery, New York, Churchill Livingstone, 198.5, p 23:3
1:3. Fullarton GM, Birnie CG, \laedonald A, et al: Controlled trial of heater probe treatment
in bleeding peptic ulcers. Br J Surg 76:541, 191)9
14. Hay JM, Lacaine F, Kohlmann C, et al: Immediate definitive surgery f()r peri,)rated
duodenal ulcer does not increase operative mortality: A prospective controlled trial.
World J Surg 12:705, 1988
],5, Hprringtoll JL, Bluett MK: The surgical management of recurrent ulceration, Conternp
Surg 28:1.5, 191)6
16, Herri,;gton JL, Daddson J: Bleeding gastroduodenal ulcers: Choice of operations, \Vorld
J Surg 11::304, 191)7
17, Hoflil1'UlI1 L Devantier A, Koelle T, et al: Parietal cell vagotomv as an emergency
procedure for bleeding peptic ulcer. Ann Surg 206:.58:3, 19S7
11). Hoffmann J, Shokouh-Amiri MH: Treatment of recurrent ulceration after vagotomy flll'
duodenal ulcer. Acta Chir Scam! 547(suppll:82, 1988
19, Hubert JP, Kiernan PD, \Velch JS, et al: The surgical management of hleeding stress
ulcers. Ann Surg 191:672, 1980
20, Hunt PS: Bleeding gastroduodenal ulcers: selection of patients for surgery, \Vorld J Surg
11:2H9, 19S7
21. Jamicson GG: Should we be changing our operative strategies for the acute complications
of peptic ulcer disease? Aust NZ J Surg .58:.525, 1988
22, Jensen 1\10, Buhrick MP, Onstad GH, et al: Changes in the surgical treatment of acid
peptic disease. Am Surg 51:.5,56, 198.5
2:3. Johnston D: Duodenal and gastric ulcer. In Schwartz SI, Ellis H (eds): :Vlaingot"s
Abdominal Operations, cd H, '\iorwalk, Appleton-Century-Crofts, 1985, p 741
24, Johns(on D: Operative technique of highly selective (parietal cell) vagotomy, Acta Chir
Scand 547(suppl):49, 191)tl
2.5. Johnstoll D, B1ackett HL: Hecurrent peptic ulcer. World J Surg 11:274, 191)7
26, Johnston D, Lyndon PJ, Smith HB, et al: Highlv selective vagotomy without a drainage
procedure in the treatment of haemorrhage, perforation, amI pyloric stenosis due to
peptic ulcer. Br J Surg 60:790, 1973
27, Johnston JH: The sentinel clot and invisible vessel: Pathologic anatomy of hleeding peptic
ulcer (editorial), Gastrointest Endosc :30::31:3, 19H4
2S. Johnston Jll, Sones JQ, Long BW, et al: Comparison of heater probe and YAG laser in
endoscopic treatment of major bleeding from peptic ulcers, Gastrointest Endose :31:17.5,
191),5
29. Kennedy 1', Connell AM, Love AHG, et al: Selective or truncal vagotomy? Br J Surg
60:944, 1973
:30. King 1''>1, Hoss AHMcL: Perforated duodenal ulcer: Long-term results of omental patch
closure, J Royal Coli Surg Edin :32:79, 191)7
:31. Koness RJ, Cutitar M, Bmchard K\V: Perf<lrated peptic ulcer: Determinants of morbidity
and mortalitv, Am Surg .56:21)0, 1990
32, Kozarek RA: Hvdrostatic-balloon dilation of gastrointestinal stenoses: A national survey,
Gastrointest Endosc :32:15, 1986
3:3. Kurata JH, Corboy ED: Current peptic ulcer time trends, J Clin Castroentcrol 10:259,
1981)
1012
AJIT
K.
SACIIDEYA ET AL.
:34. Lanng C, Palnaes Hansen C, Christensen A, et al: Perf(mlted gastric ulcer. Br J Surg
7.5:7.58, 1988
3.5. Larson G, SC'lllnidt T, Gott J, et al: Upper gastrointestinal bleeding: Predictors of
outcome. Surgery 100:76.5, 1986
36. Laurence BH, Cotton PB: Bleeding gastroduodenal ulcers: Nonoperative treatment.
World J Surg 11:295, 1987
37. Lunde 0, Liavag I, Roland M: Proximal gastric vagotomy and pyloroplasty for duodenal
ulcer with pyloric stenosis: A thirteen-year experience. World J Surg 9:165, 198.5
38. McConnell DB, Baba GC, Deveney CW: Changes in surgical treatment of peptic ulcer
disease within a veterans hospital in the 1970s and the 1980s. Arch Surg 124:1164, 1989
39. McGee GS, Sawyers JL: Perforated gastric ulcers. Arch Surg 122:5.5.5, 1987
40. ~Iulholland MW, Dellas HT: Chronic duodenal and gastric ulcer. Surg Clin North Am
67:489, 1987
41. Hheault ~IJ, Legros G, Nyhus L: Reflux gastritis. Dig Surg 5:5, 1988
42. Rogers PN, Murrav WH, Shaw R, et al: Surgical management of bleeding gastric
ulceration. Br J Surg 75:16, 1988
43. Roher HD, Thon K: Impact of early operation on the mortality from bleeding peptic
ulccr. Dig Surg 1:32, 1984
44. Schinner BD: Current status of proximal gastric vagotomy. Ann Surg 209:131, 1989
4,5. Simon B, Porro GB, Cremer M, et al: A single nighttime dose of ranitidine 300 mg versus
ranitidine 1.50 mg twice daily in the acute treatment of duodenal ulcer: A European
multicenter trial. J Clin Gastroenterol 8:367, 1986
46. Sontag S: Current status of maintenance therapy in peptic ulcer disease. Am J Gastroenterol 83:607, 1988
47. Stabile BE, Passaro E Jr: Progress in gastroenterology: Recurrent peptic ulcer. Gastroenterology 70: 124, 1976
48. Steele RJC: Endoscopic haemostasis for non-variceal upper gastrointestinal haemorrhage
(review). Br J Surg 76:219, 1989
49. Thirlhy HC, Feldman ]\\: Transthoracic vagotomy for postoperative peptic ulcer. Ann
Surg 201 :648, 198.5
50. Thompson JC: The stomach and duodenum. In Sabiston DC (ed): Textbook of Surgery.
Philadelphia, WB Saul1l1ers, 1986, p 810
.51. Thurston OG: Perforated benign gastric ulcer: The case for simple closure. Can J Surg
32:94, 1989
52. Toftgaard C: Gastric cancer after peptic ulcer surgery: A historic prospective cohort
investigation. Ann Surg 210: 159, 1989
.53. Turner WW, Thompson WM, Thai EH: Perforated gastric ulcers. Arch Surg 123:960,
1988
.54. Walker LG: Trends in the surgical management of duodenal ulcer. Am J Surg 1.5.5:436,
1988
.55. Wastell C: Proximal gastric vagotomy. In Nyhus LM, Wastell C (eds): Surgery of the
Stomach and Duodenum, ed 4. Boston, Little Brown and Company, 1986, p 36.5
56. Zuckerman GH, Shuman H: Therapeutic goals and treatment options for prevention of
stress ulcer syndrome. Am J ~Ied 83:29, 1987
Address reprint requests to
Ajit K. Sachdeva, MD
Department of Surgery
The Medical College of Pennsylvania
3300 Henrv Avenue
Philadelphia, PA 19129